Abstract
We conducted a cross-sectional, survey study of 764 volunteers to gain insight into patients’ perceptions of physician qualities of compassion and competence. Among 651 (85% response rate) survey participants, mean age was 52.4 (SD 21.4) years, 70.8% (n = 458) were female, and 84% (n = 539) identified as white. Predictors of compassion over competence included female gender (adjusted odds ratio [aOR] = 1.4, 95% CI: 1.04-1.89) and whether the respondent had a personal connection to the vignette (aOR = 1.24, 95% CI: 1.0-1.53). Thematic analysis demonstrated that preferences were influenced by: (a) explicit beliefs regarding the value of physician compassion and physician competence; (b) impact of emotional and mental health on medical experiences; (c) the type and frequency of health care exposure; and (d) perceived role of the physician in various clinical vignettes. Patients had wide-ranging, complex opinions on the qualities they valued in their physicians. These findings suggest that patients are engaged and can provide critical thoughtful feedback on the practice and delivery of health care.
Keywords
Introduction
Physicians must navigate overbooked schedules, ever-evolving electronic medical record systems, and meeting institutional and national quality metrics all within strict time constraints. When coupled with the emotional exhaustion that often accompanies caring for others, many may inadvertently find their compassion skills declining secondary to burnout and compassion fatigue (1 -3). Although the key to providing adequate care may be ensuring medical competency, several studies demonstrate that compassion has the ability to improve outcomes and patient satisfaction (4 -7). There is also evidence that a unique interplay between these two traits exist, as those physicians perceived as more compassionate by patients are also viewed as more competent. Furthermore, being competent in a particular skill or procedure requires time and repetition, which can come at the cost of compassion. This is complicated by the fact that physicians’ opinions of their own compassion are often discordant with their patients’ perceptions (8 -14). Since there is clear evidence demonstrating the value of physician compassion, there is a need for investigation into more effective and efficient ways of compassionate care delivery.
One way to optimize compassion delivery is for physicians to take a more targeted approach. Rather than striving to always be perceived as compassionate and competent, is there a way they can predict the appropriate approach to patients based on the different clinical or demographic characteristics? If so, perhaps physicians can refocus their efforts to address the most important needs of their patients. The goals would be to assess and fulfill these important patient needs and expectations, increase patient satisfaction, and achieve better outcomes. For example, patients routinely rank intrinsic provider characteristics such as compassion equally, if not more important, than technical skills when caring for patients in end-of-life care; however, the preferences of patients in other scenarios such as surgical interventions and chronic disease management are not well characterized (15 -17).
In this study, we sought to gain insight into patients’ expectations in a variety of clinical vignettes. Specifically, we hypothesize that compassion is preferred less than competence in hypothetical vignettes with higher acuity, such as in a surgical context, and may vary with certain patient characteristics.
Methods
Survey Development
The survey was developed by two team members (K.H., L.M.). The survey design was reviewed by all team members and revised following feedback from adult patient volunteers from the Office of Patient Experience of the University of Michigan and the University of Michigan Geriatrics Center patient registry. Reviewers provided feedback regarding the wording, appropriateness, and comprehensibility of vignettes. The patient volunteer reviewers were not part of the final study.
Survey Design
The survey was comprised of 7 clinical vignettes, followed by both closed- and open-ended questions regarding the importance of physician compassion and competency (Supplemental text). We defined compassion as the physician’s ability to be sympathetic, kind, and sensitive to the needs of patients. Competency was defined as the physician’s knowledge, skills, and abilities to provide good medical treatment. In 5 of the 7 vignettes, participants were asked to place themselves in the role of the patient as they navigate several clinical scenarios, including surgical care (e.g., consideration of cataract surgery and undergoing brain surgery, and developing subsequent complications), long-term care (e.g., chronic management of type 2 diabetes mellitus with unique lifestyle considerations, and heart failure management encompassing end-stage disease), and end-of-life care (e.g., terminal lung cancer). The remaining 2 vignettes were pediatric cases where participants were asked to place themselves in the role of the patient’s parent or guardian (e.g., colonoscopy and bone marrow biopsy).
Following each clinical vignette, participants were asked to rank the importance of physician compassion and competency on a 5-point Likert scale (1: unimportant, 5: essential). They were then asked which quality (compassion or competence) they would choose if they could only select one and to provide a brief explanation of their choice. Lastly, participants were asked to note whether they have any significant personal connection to the vignette, either through their own experiences, their family, or through close friends.
Demographic information including age, gender, race/ethnicity, education level, income level, number of children, and whether they have or had a career in health care were collected. We also collected information reflective of respondent health, including how many times they see a primary care or specialty physician each year, how many hospitalizations they had in the past 2 years, and their perceived general health status (e.g., poor, fair, good, very good, or excellent).
Study Population and Design
We sent the survey to 800 individuals: 618 are part of a patient registry through the University of Michigan Geriatrics Center who have indicated that they are interested in participating in research studies; 182 are in the University of Michigan Health Research Database and volunteered to participate in our study. The survey was sent once via post-mail, along with a US$5 incentive and a self-addressed stamped envelope. No follow-up letters to nonrespondents were sent. The study is deemed institutional review board (IRB)-exempt by the IRB of the University of Michigan.
Data Analysis
Quantitative data
Survey responses were abstracted, coded, and entered into a database by trained research personnel (K.H., K.G., B.L.). STATA version 13 was used to analyze the data. Ordinal logistic regression was performed to determine the predictive effect of demographic factors.
Qualitative data
Free-text responses for each vignette were reviewed and assessed for major themes relevant across all cases and subthemes pertinent to each case individually. A coding scheme based on the major themes and subthemes was developed. Two investigators (C.A.V., K.H.) independently coded 200 responses from each case or until thematic saturation was reached. Inter-rater reliability was established through an iterative process of consensus building between the two researchers (C.A.V., K.H.). All authors then met to discuss and establish the themes and subthemes. Representative quotes for each major theme were identified and compiled for illustrative purposes.
Results
Respondent Characteristics
Of 800 surveys mailed, 36 were returned due to incorrect or change in address, and 651 surveys were completed and returned (651/764 = 85.2% response rate to a one-time mailing). Of the responses, 19% were received within the first week, and 92% were received within the first month. Respondent demographics are displayed in Table 1. The mean age of respondents was 52.4 (SD 21.4 years) years, 70.8% (n = 478) were female, and 84.0% (n = 539) were non-Hispanic white. A majority (73.5%, n = 474) reported having an undergraduate college degree or higher. Approximately one-fourth (25.4%, n = 163) reported currently working or having worked in health care. Annual income level was well-dispersed, with the highest percentage of respondents (44.7%, n = 277) reporting <US$50 000. Approximately 68.3% (n = 439) reported that their health is excellent or very good; 31.7% (n = 204) report having fair/poor health. Most respondents (58%, n = 374) reported seeing their physician 2 to 5 times/year, and 71.2% (n = 460) reported that they have not been hospitalized in the past 2 years.
Respondent Demographic Characteristics.
Preferences for a Compassionate or Competent Physician
Respondent preferences for a compassionate or competent physician varied by vignette (Figure 1). In general, a greater number demonstrated a preference for competence over compassion as the acuity or risk increased. For instance, in the surgical and pediatric vignettes, where medical acuity or perceived risk was high, physician competence was preferred over physician compassion. Specifically, when choosing a physician for planned cataract surgery, with a risk of vision loss, 93.4% chose competence over compassion (2.5%, P < .001). Similarly, 89.4% chose competence over compassion (4.4%, P < .001) when being evaluated for brain surgery. However, when a complication from brain surgery was introduced, there was a statistically significant increase in a preference for physician compassion (4.4%-23.4%, P < .001).

Choice of compassion or competence in each clinical vignette.1* Competence preferred (P < .001). † Compassion preferred (P < .001). ‡ No significant preference (P = .375).1Two of the clinical vignettes (about brain surgery and heart failure) had 2 parts, referred to in Figure 1 as part A and part B.
Predictors of compassion over competence include female gender (adjusted odds ratio [aOR] = 1.4, 95% CI: 1.04-1.89) and whether a respondent has a personal connection to the vignette (aOR = 1.24, 95% CI: 1.0-1.53) across all vignettes. In contrast, younger age is associated with less likelihood of choosing physician compassion over physician competence (aOR = 0.93, 95% CI: 0.89-0.97). Race, education, a career in health care, income, health status, and number of physician visits and hospitalizations in the past year and 2 years are not associated with preferences for compassion or competence (Table 2).
Multivariable Regression Models of Choice for Compassion Over Competence.
Abbreviation: aOR, adjusted odds ratio.
Factors Influencing Respondent Preferences
Thematic analysis of the open-ended responses demonstrated that respondent preferences are influenced by four major themes (Supplemental Table 1): (a) explicit beliefs regarding the value of physician compassion and physician competence; (b) impact of emotional and mental health on medical experiences; (c) the type and frequency of health care exposure; and (d) perceived role of the physician in various clinical vignettes.
Explicit beliefs regarding the value of compassion and competence
In surgical vignettes where physician competence was more often preferred, respondents reflect on the importance of a physician’s technical skills, the fear of permanent consequences resulting from incompetence, and that perhaps compassion was not the physician’s role, and instead may be provided by other members of the care team. Similar sentiments were shared in the pediatric cases where physician competence was preferred, but physician compassion was very important. In contrast, the end-of-life vignette where compassion was most often preferred, respondents noted the importance of making the patient and family feel cared for and that the mental and psychosocial health of the patient became a higher priority. In chronic care vignettes, preferences for physician competence and compassion varied. Respondents reflected on the importance of the physician’s ability to provide counsel on both medical management and lifestyle changes, while also understanding the patient’s limitations, adjusting therapeutic regimens, and modifying expectations of compliance accordingly.
Even in vignettes where the majority of respondents preferred competence (e.g., surgery) or compassion (e.g., end of life), a distinct role for both qualities remains (Table 3). Among surgical vignettes, respondents often noted that although physician competence is the priority, a physician’s compassion had a substantial impact on patient entrustment. Respect for the patient’s autonomy and decision-making abilities was also important, as was their perception that the physician actually cared about them and would act in their best interest should a complication arise. Furthermore, respondents often reflected on the importance of a compassionate approach once competency had been established, suggesting a complex interplay between these two qualities. In the end-of-life vignette, respondents often reported that although physician compassion was prioritized, physician competence was important, and impacted their confidence in the physician’s accurate diagnosis and prognosis. Additionally, establishing physician competence may provide some reassurance that all avenues of treatment had been explored, that the physician was open to considering other treatment options, and that they would support the patient and their family even at the end of life.
Role of Compassion in a Scenario Where Competency Was Scored High (Surgery) and Role of Competence in a Scenario Where Compassion Was Scored High (End of Life).
Impact of emotional and mental health on medical experiences
Table 4 displays quotes on how respondents define and choose compassion or competence and the influence their emotional health had on their decisions. Respondents’ desired outcomes were often similar between vignettes, but the ways in which they expected the situation to be approached were very different. Among surgical vignettes, a preference for physician competence was often influenced by the desire for confidence in the physician’s skill and comfort knowing they are in capable hands. A preference for physician compassion was also influenced by a respondent’s desire for confidence and trust in their physician, but some responded that this could only be established once the physician demonstrated competence. Similarly, in pediatric vignettes, a preference for physician competence was influenced by the respondent’s desire for confidence in their skill and ability, whereas a preference for physician compassion was often associated with the desire for reassurance and respect for the emotional challenges the parent or guardian and child experience, particularly in serious conditions such as pediatric cancers.
Impact of Emotional and Mental Health Across Vignettes.
Type and frequency of health care exposure
Physician compassion was identified as key when subjecting a child to painful tests or procedures, particularly when considering the long-term consequences of physician interactions on the child’s health. In the end-of-life vignette, many respondents noted that mental and emotional health becomes a higher priority when dying, thus suggesting a preference for physician compassion. Others noted that they would prefer less focus on their emotional well-being and greater focus on understanding the technical complexities of their illness and the impact on prognosis. Among vignettes dealing with the chronic disease management, a preference for physician competence was often influenced by the respondent’s desire for confidence in the physician’s knowledge, whereas a preference for physician compassion was influenced by a desire for emotional support at diagnosis and to ensure long-term treatment compliance and resilience.
Perceived role of the physician in various clinical vignettes
A desire for a more compassionate or competent approach to care was reflective of the different roles participants looked to their physician to fulfill. Although physician competence was vital, it was insufficient in providing “effective” medical care and must be used in conjunction with a more compassionate approach. As reflected in the responses, compassion is multifaceted and utilized in a variety of ways depending on the scenario and the physician’s roles.
The perceived role of the physician was discussed extensively by respondents in the various clinical vignettes. Six roles emerge with regard to the physician–patient relationship: healer, advocate, counselor, friend, educator, and enforcer. Table 5 displays how participants defined the specific roles and their reflective quotes.
Perceived Role of a Physician.
Although patient preferences regarding the role of the physician varied greatly, many participants preferred to engage in a more team-based approach to managing their health. In addition to a healer and educator, the respondents looked to their physician to fulfill the role of a counselor; someone who provides compassionate guidance as they navigate individualized treatments and lifestyle changes. This role may grow overtime as patients began to view their physician as a friend with whom they may celebrate their successes, vent their frustrations, and disclose their failures. Conversely, there were also individuals who desired the more traditional approach to care. Respondents reported personal responsibility or blame for their health conditions and looked to a physician to provide strict instruction and goals they should try to achieve as a way to improve health long term. In the event that someone failed to meet these goals, the respondents preferred their physician to take on the role of an enforcer by conveying realistic depictions to motivate them into making changes.
Discussion
We used a survey-based study to explore patient perspectives of physician compassion and competence by asking respondents to reflect on the importance of these qualities across a variety of clinical vignettes. The responses clearly indicate that patients want to be heard and actively engaged in discussions on how their medical care is approached. Respondents generally value physician compassion for those vignettes pertaining to end-of-life care and chronic disease management, and physician competence particularly in procedure-oriented vignettes. We find that specific factors influence patient preferences for one characteristic over another.
As the acuity of care increased, respondents often reported a desire for both a technically proficient physician and one who provides enough information and moral support for them to make informed decisions about their care. In addition, the physician’s ability to instill a sense of compassion and empathy was key to improving clinical outcomes. This is consistent with several studies demonstrating that in trauma settings, patients who rank their physicians as having lower empathy have also been shown to have lower subjective evaluations of their surgical outcomes (18 -20).
Good clinical outcomes in chronic disease management are dependent upon patient compliance with treatment and lifestyle choices and changes. The goals of care expand beyond medical management to include patient empowerment, establishment of ongoing trust, and patient retention, in order to enhance compliance to medications and other treatment and thus manage a chronic condition for prolonged periods of time. To achieve these goals, establishing a solid physician–patient relationship is critical (21,22). As evidenced by the narratives, patients facing chronic care recognize this. However, the nature of this relationship and the perceived role of the physician in these settings vary substantially among individuals, requiring the physician to modify their approach based on the type and severity of the condition. However, more investigation is needed to determine what drives patient preferences and how best to incorporate this into communication education and techniques for providers.
Several limitations should be considered. First, perceptions of what constitutes a compassionate or competent physician can vary from person to person, as well as their individual preferences for approach regardless of clinical condition as influenced by prior experiences or personal beliefs. Although the purpose of this study was not to define compassion or competence from the patient’s perspective, our goal was to better understand unique patient preferences for communication and approach to medical care across a diverse range of clinical vignettes. With this study—the largest study examining patient preferences across a diverse range of clinical vignettes—our intent was to go beyond the traditional and very brief patient satisfaction questions. Second, although we asked questions about prior exposure to health care, future work should align patient preferences with objective outcomes such as prolonging life, successful surgery without complications, or subjective outcomes such as the abilities of the physician to elicit patient concerns effectively and accurately. Future studies should also explore institutional considerations such as physician evaluations, reimbursements, promotions, job security, and clinical competencies gained.
Patients want a competent physician, but compassion plays an important role. Time remains a major factor in achieving both that would provide superb care, achieve institutional goals, and enhance both patient and physician satisfaction. Our findings point to the inherent limitations within the current system of health care delivery. Physicians are increasingly scrutinized by their billing, completeness in documentation, and meeting quality indicators, which may lead to moral distress and burnout. If we want more people to enter the medical field, it is imperative that we develop systems to reward or allow compassionate care more readily. Such improvements will undoubtedly enhance quality, patient satisfaction, and improve outcomes, which in turn will play a significant role in reducing health care costs.
Supplemental Material
Supplemental Material, CC_Supplement_7.16.20 - A Survey of Patient Perspectives on Approach to Health Care: Focus on Physician Competency and Compassion
Supplemental Material, CC_Supplement_7.16.20 for A Survey of Patient Perspectives on Approach to Health Care: Focus on Physician Competency and Compassion by Kevin Heinze, Pasithorn A Suwanabol, C Ann Vitous, Paul Abrahamse, Kristen Gibson, Bonnie Lansing and Lona Mody in Journal of Patient Experience
Footnotes
Authors’ Note
Kevin Heinze, MD, and Pasithorn A. Suwanabol, MD, MS, contributed equally to this work. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.
Acknowledgments
The authors thank the members of the Michigan Medicine Patient and Family Advisory Council for their input on the vignettes and survey design. The authors also acknowledge the Michigan Institute for Clinical and Health Research (MICHR)
registry as well as thank the survey respondents for their participation and thoughtful responses.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by the National Institute on Aging (LM, K24 AG050685) and the Michigan Institute for Clinical and Health Research (LM, UL1TR002240). PAS receives funding through the National Institute on Aging (R03 AG056588) and the Agency for Healthcare Research and Quality (K08 HS026772).
Supplemental Material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
